Logo IMG
HOME > PAST ISSUE > March-April 2014 > Article Detail


The Challenge of Manufacturing Between Macro and Micro

Classic ways of folding paper into dynamic shapes—origami, pop-up books—inspire methods to engineer millimeter-scale machines.

Robert J. Wood

Scaling Down

2014-03WoodFp126bot.jpgClick to Enlarge ImageTo get to the meso-scale, we have to first understand how a device will operate as its size is reduced. A typical exercise involves analyzing the forces it will experience. For example, when all dimensions scale equally, if one dimension decreases by a factor of 10, volume decreases by 1,000. Therefore, the force from gravity (which is related to mass) is also reduced by 1,000. However, the force from friction with surrounding objects will only decrease by a factor of 100 because it is dependent on area. As objects get smaller, therefore, surface forces such as friction begin to matter much more than gravity does. A consequence is that, say, the kinds of rotary bearings used to create smooth motion in axles and wheels become more and more inefficient due to friction as the scale is reduced. The combination of reduced efficiency and the challenges of manufacturing at small scales has motivated alternatives for movement that are instead based on bending, like hinges. A similar scaling analysis can be done for fluid forces, electromagnetic forces, electrostatic forces, and surface tension.

2014-03WoodFp126top.jpgClick to Enlarge ImageJust as forces and effective mechanisms change with scale, so do practical manufacturing methods. To create meso-scale devices, we first examined the viability of scaling down existing macro-scale processes, and the combination of some of these methods into new processes.

Manufacturing methods that create structures by depositing or adding layers of material are called additive processes. One prominent example is 3D printing, and it has evolved tremendously over the past few decades in both research and commercial environments. Commercially available printers use a variety of materials and methods, from the setting of heated wax to inkjet-style printers that extrude ultraviolet- curable polymers, to stereolithography systems (which uses an ultraviolet laser to cure layers of liquid resin) and lasers that selectively sinter powder into a solid form. Such systems have dramatically reduced the barriers to entry for generating three-dimensional parts—and even articulated mechanisms can be created in one print without assembly.

The primary drawbacks for commercial 3D printers are the cost of raw materials and material quality. Most printers are limited to either a single material or a single type of material (usually polymers). Regardless of these limitations, 3D printers have achieved widespread adoption both for prototype development in industry and as an invaluable educational tool in academia. More recently, the “maker” movement—a subculture that encourages everyday people to invent and create—has democratized access to 3D printers by creating low-cost versions of the devices. 3D printers are nearly affordable for home use and are following a similar price and adoption trend as the now ubiquitous 2D printers in nearly every home and office.

Research in 3D printing has focused on expanding the variety of printable materials and exploring the limits of scalability. Expanding material diversity is a critical step toward monolithic printed devices—ones that include all their components without any additional assembly. New high- performance polymers, printed conductors, and even complete batteries have been demonstrated recently. With respect to scaling, one example of commercially available micro-scale 3D printing is two-photon lithography, which uses lasers to directly write 3D patterns in a polymer. Submicron resolution is possible, but at the expense of build time for structures with centimeter or larger size scales.

Alternatives to additive manufacturing where materials are directly deposited include molding and casting. Such processes have benefits for inexpensive mass-produced components, yet for submillimeter scale, they still require specialized tooling to create the mold. One exception is soft lithography, which grew out of research in microfluidics, where microscopic channels are used to transport tiny quantities of fluid, the basis for devices such as inkjet printers and DNA-analyzing chips. Soft lithography leverages a range of tools borrowed from integrated circuit fabrication, with an added emphasis on the use of surface properties to create layered devices. First the integrated-circuit technique of photolithography is used to pattern a photoresist—for example, a pattern in a layer of material is cured by light and the uncured sections are chemically etched away. In this case the photoresist is used as a mold to create features in a flexible polymer or as a stamp for subsequent contact-based material deposition.

The process is inherently cheap: It requires a few tens of dollars worth of materials, a transparency mask, minimal (and common) lab equipment, and modest specifications for cleanliness (compared to integrated circuit production). However, soft lithography results in quasi-two-dimensional devices; to construct 3D structures requires methods for aligning and bonding multiple layers, which could increase cost, complexity, and process time. Also, for any additive process that builds structures layer by layer, position errors in each of the deposition iterations will integrate over the course of fabrication. This concern requires either high-fidelity position feedback, or the device must be able to withstand increased errors as the number of layers in a build is increased.


comments powered by Disqus


Subscribe to American Scientist